June 2019
1
Security and justice evidence mapping update
June 2019
Paul Jackson, Joe Bell and Shivit Bakrania
June 2019
2
About this report This report was prepared for the UK Department for International Development. The views
expressed in this report are those of the authors, and do not necessarily reflect the opinions of
GSDRC, its partner agencies or DFID.
Acknowledgments GSDRC would like to thank Mahek Toosy and Ravina Ishtiaq for additional work on screening and
Joanne Dolan for work on screening and coding of the material. The authors would also like to thank
Brian Lucas (GSDRC) for his substantial input throughout the project, particularly on the database. In
addition, several experts contributed to the project, including Rachel Kleinfeld & Tom Carothers,
CEIP, Adrian Di Giovanni, IDRC, Pilar Domingo & Lisa Denney, ODI, Robert Muggah, Igarapé Institute,
and Erwin van Veen, Clingendael.
About GSDRC: GSDRC is a partnership of research institutes, think-tanks and consultancy
organisations with expertise in governance, social development, humanitarian and conflict issues.
We provide applied knowledge services on demand and online. Our specialist research team
supports a range of international development agencies, synthesising the latest evidence and expert
thinking to inform policy and practice.
GSDRC, International Development Department, College of Social Sciences University of
Birmingham, B15 2TT, UK
www.gsdrc.org; [email protected]
June 2019
3
Contents
1 Introduction 7
2 Scope 7 2.1 Thematic Scope 7 2.2. Intervention, Outcome and Output Categories 8 2.3 New and Expanded Areas
2.4 Inclusion Criteria 9
11
3 Methodology 12 3.1 Search Strategy 12 3.2 Screening 13 3.3. Coding 14 3.4 Quality Control 14
4 Search Results 14
5 Features of the Evidence Base 16 5.1 Research Design 16 5.2 Publication Form 17 5.3 Publisher 17 5.4 Geographic Scope 17 5.5 Quality of Evidence 18
6 Evidence Map Gap 18 6.1 Overview 19 6.2 Analysis: Interventions 24 6.3 Analysis: Outputs 25 6.4 Analysis: Outcomes 26
7 What has Changed? 28 7.1 Areas and Significant Change 28 7.2 Areas Related to Expansion of Search Terms 28 7.3 Detailed Changes 29
8 Potential Avenues for Future Research 30 8.1 Areas Suitable for Synthesis Research 31 8.2 Priority Research Areas 31
References 33
Appendix 1: Security and Justice Evidence Map Updated Research Protocol (Updated after Scoping Report)
34
June 2019
4
Executive summary This evidence mapping was conducted to identify the empirical evidence on the outcomes of security
and justice (S&J) interventions ranging from security reform to training border guards (a full list is on
page 9) and to update an original evidence mapping published in September 2015. The findings are
presented in the form of an evidence gap map (p.15), which provides an accessible and visual
representation of where the evidence for S&J programming is more abundant or limited.
The evidence mapping included a rigorous review of the existing S&J evidence base and identification
of new evidence and material in an expanded search since 2015. Selected sources, including journal
indices, online research and evaluation repositories, resource centres and experts were interrogated.
Studies that explored interventions and results were selected for inclusion in a database (available to
download) and coded according to the publication type, thematic focus, intervention, output and
outcome categories.
Features of the evidence base Research design: The majority of studies continue to be mostly observational in nature, with only a
1% differential compared to 2015 (94%). Most of these used the same techniques (interviews, focus
groups, ethnography, historical analysis and political economy analysis). Fifteen experimental or
quasi-experimental studies were identified, an increase on the eight from 2015, and twenty-eight
secondary studies (up from fifteen) covering a mixture of literature reviews, and annotated
bibliographies. Having mainly observational case studies means a lack of cross-country comparisons,
but a depth of knowledge about specific countries. Additionally, the very small number of
experimental studies, means there is limited information on the effectiveness of S&J interventions.
Publication form: The largest group (46%) of documents remain peer-reviewed journal articles. Thirty-
five per cent of studies are classified as ‘other’ reports; these include analytical reports and case
studies, and were mostly published by academic organisations, think tanks and non-governmental
organisations. Whilst few evaluations (12%) continued to meet the inclusion criteria, this is double
(6%) the previous 2015 map. Those that did meet the criteria include thematic evaluations of donor
security and justice programming, synthesis programme evaluations and other individual programme
or country evaluations. This suggests that evaluations may be improving in quality, but there is scope
to improve this further.
Geographic scope: This has not materially changed since the 2015 map. Western Africa is by far the
most studied region. South Asia and Eastern Africa are the next most studied regions. There are very
few studies exploring Central Asia, East Asia, Eastern Europe and Northern Africa.
Thematic focus: Policing remains by far the most studied theme, followed by access to justice/legal
empowerment, justice sector reform, legal reform and non-state actors. Individual studies often
explore interventions that addressed more than one theme/sector. The evidence base on
interventions aimed at, or including, non-state actors showed a relatively large increase. This seems
to reflect an increased role for non-state actors in approaches like community policing and local
justice. There is still a lack of evidence in some important areas like intelligence and border security.
Within the ‘access to justice/legal empowerment’ theme most of the existing evidence is for the
former and there is very little evidence on actual legal empowerment.
Gap map analysis Evidence gaps and areas of strength appear where one would expect them, and more or less in line
with the 2015 analysis. The gaps appear where the outputs and outcomes are less tangible, and more
difficult to operationalise and measure (section 6). The evidence is scarce for a number of output and
June 2019
5
outcome categories that are justice-specific, whilst stronger evidence can be found on outcomes and
outputs that are more general and applicable across a wider range of intervention types. Detailed
definitions for each of the intervention, output and outcome categories are tabulated on pages 45-
51. In terms of the evidence, ‘abundant’ is coded in green on the tables on pages 21-23, whereas
‘limited’ is coded as red.
Short- and medium-term results, which are tangible, operational and directly related to intervention
categories, are labelled ‘outputs’, while longer-term and more indirect results are labelled ‘outcomes’.
The distinctions were based on DFID’s Security Sector Reform (SSR) Theory of Change (ToC).
Interventions
• The evidence is abundant for: capacity building of organisations; strategic/statutory
frameworks and legislation; community-based approaches; and restructuring of the security
and justice (S&J) sector. The evidence base on non-state actors has also improved since 2015.
Many studies provide examples in which a combination of these interventions is used.
• The evidence is limited for: preventative interventions (integrated efforts to prevent violence
and crime); Disarmament Demobilisation and Reintegration (DDR); and integrated political
engagement/activities that promote political will. Evidence is also scarce for
displaced/refugee interventions. There were zero impact evaluations of mutual legal
assistance. This was discussed during the research period because it subsequently does not
show up in the database but is clearly a gap.
Outputs
There have been no significant changes in the balance of where evidence is focused since 2015:
• The evidence is abundant for: the capacity of state and non-state organisations to deliver;
strategic frameworks; confidence, trust or satisfaction on the part of citizens in providers; and
roles, coordination and dialogue amongst organisations and agencies. These are considered
to be key outputs for S&J programming, they are relatively tangible, and are directly related
to intervention categories.
• The evidence is limited for: application, compliance with and interpretation of laws; and
state/non-state linkages. These are both justice-specific outputs.
Outcomes
• The evidence is abundant for: ownership of reforms by national and local stakeholders;
resource allocation/funding stability and sustainability; access to provision; stability and
outbreaks of conflict or violence; and human rights measures implemented to improve
compliance. These outcomes are considered core tenets of S&J and/or are relatively tangible.
• The evidence is limited for: incentives for improved service delivery amongst security and
justice actors; actual crime rates; legal awareness and confidence; gender-based violence
rates; and judicial redress to protect rights. The evidence is particularly limited for longer-term
development outcomes such as: economic development (local or national); poverty
reduction; access to land, inheritance and property rights; access to public services and
economic resources.
June 2019
6
Potential avenues for future research Synthesis research focusing on those themes and interventions for which there is more evidence may
help build a more rigorous evidence base. Importantly, there are still very few experimental or quasi-
experimental studies in the database and thus analysis on the causal linkages between interventions
and outputs or outcomes is lacking. Increasing the number of experimental, and even quantitative
observational, studies could improve understanding of which interventions are effective.
Overall the results of the update to the mapping indicate only very limited changes in the thematic
focuses, except in some specific areas, notably non-state actors and gender analysis, where there has
been growth. The updated map has expanded the scope of the search to address some of the gaps
identified in the 2015 map, although the overall findings remain similar.
• The evidence base is not as large for the ‘developing accountability’ intervention, but there is
abundant evidence on the accountability output. One might have expected the depth of
evidence for these directly related interventions and outputs to be similar. Studies comment
on the impact of different types of interventions on accountability as a normative measure of
success, even when interventions to develop accountability are not explicitly mentioned.
• The evidence for ‘integrated political engagement/activities that promote political will’
remains much smaller than for the other pre-determined intervention types, even though it
is acknowledged more often. Studies generally do not discuss these types of activities, even if
they might be happening in the background of other more tangible interventions. There is
limited evidence that political engagement/political will interventions can affect the outcome
‘incentives for improved service delivery’, even though these would appear to be directly
related. Another anomaly is that one might expect more evidence to appear in combination
with the outcome ‘political will to enact reforms’.
• The 2015 map found that there was limited evidence for some types of intervention where
one might expect the evidence base to be larger. In 2015 this included available evidence on
gender-specific interventions. The 2019 map finds that this has improved considerably and is
no longer such a gap, though there remains very little good quality data related to gender-
based violence rates.
• Non-state actors is another area where the 2015 map found very little but the evidence in this
area has significantly improved and is no longer a gap overall. However, the evidence base is
still relatively small on the outcome of access to land, inheritance and property rights.
• Evidence on preventative interventions remains limited across all categories. One might have
expected more evidence for the outcomes: stability and outbreaks of conflict or violence;
actual crime rates; and gender-based violence rates.
• A significant area where there is a need for more research is around issues of migration,
displacement and refugees.
June 2019
7
1. Introduction This rigorous evidence mapping exercise draws upon the existing security and justice (S&J) evidence
map (https://gsdrc.org/publications/security-and-justice-evidence-mapping/) completed in 2015,
bringing it up to date and extending it to cover new subjects of interest. The ultimate objective is to
produce an evidence gap map, which illustrates where there is existing evidence on S&J programming
and where there are gaps in the evidence base. The map does not provide information on what the
evidence says, nor does it comment on the nature of linkages between interventions and outputs or
outcomes. The map will provide a catalogue of reliable, rigorous evidence that DFID can draw upon to
inform current thinking about how S&J programming can contribute towards reaching Sustainable
Development Goal 16 (‘Peace, Justice, and Strong Institutions’).
This project built upon the existing catalogue compiled in 2015, retaining the existing data within it,
by completing the following tasks:
• Checked and updated web links to all items already catalogued (386 items).
• Adjusted and tested the search keywords used to identify the literature that will be screened for inclusion.
• Where the search strings have not changed (see protocol in appendix), the project identified and catalogued new materials published for the period 2015-2019.
• Where the search strings have changed, the project identified and catalogued new materials published for the period 2008-2019.
The existing catalogue of evidence and visual evidence map, stored in an Excel spreadsheet, has been
updated to incorporate the new material identified through this exercise and is available as a separate
document.
We maintained compatibility with the methodology used to produce the 2015 evidence map to the
fullest extent possible. A detailed description of the methodology is included in the Research protocol
in Appendix 1. Some of the sources searched in 2015 no longer exist, or have made changes to their
search engines and/or indexing methods, so the 2015 study was not replicated exactly, but
consistency was maintained as much as possible.
2. Scope The evidence mapping was based on a detailed search protocol (see Appendix 1). It included a rigorous
review of the S&J evidence base, searching a number of databases, online sources and think tanks and
other organisations. It also involved consultation with a group of experts in the area. Studies were
selected for inclusion in a database and coded according to publication type, thematic focus, and
intervention, output and outcome categories. This interactive database is available as a separate
document.
2.1 Thematic scope The search strategies focused on studies that explore interventions in the following thematic fields,
which reflect some of DFID’s core areas of interest in S&J:
• Defence
• Policing
• Intelligence
• Justice sector reform
• Access to justice and legal empowerment
June 2019
8
• Prisons
• Border security
• Non-state actors
• Legal reform
• Security promotion
• Urbanisation
• Organised crime
• Counter-terrorism
The 2015 map added border security, non-state actors, legal reform, security promotion, urbanisation,
organised crime and counter-terrorism during the coding stage. A number of limiting devices were
introduced to keep the mapping manageable. The themes ‘transitional justice’ and ‘legal reform’ were
originally included in the S&J search strategy. These were later removed in order to refine the focus
after test searches generated a very large number of results. Documents focusing on commercial
justice, juvenile and child justice, the war on drugs and drugs regulation, and environmental justice in
fragile contexts were also excluded.
There have been a number of changes to the original proposal. Non-state actors has been updated to
incorporate faith groups. Community has been split into two separate searches. The first, community,
incorporates the non-gender searches and has added spiritual leaders. The second, gender, reflects
the emphasis on gender-based approaches and gender-based violence. The emerging themes of
mutual legal assistance and displacement have been added. The category of organised crime has been
replaced by the mutual legal assistance category. The emerging theme of counter-terrorism has been
expanded to reflect the emphasis on security forces as a potential source of animosity.
2.2 Intervention, output and outcome categories
Short- and medium-term results, which are tangible, operational and directly related to intervention
categories, are labelled ‘outputs’, while longer-term and more indirect results are labelled ‘outcomes’.
The distinctions were based on DFID’s Security Sector Reform (SSR) Theory of Change (ToC)1, and
refined through further discussions with DFID advisors. The ToC assumes that outputs are more easily
measurable and tangible than outcomes and so perhaps more easily conceptualised in terms of
variables for evaluation. Detailed descriptions of specific outputs and outcomes were discussed with
advisers, external experts and researchers as part of the protocol and subsequent process. Detailed
definitions for each of the intervention, output and outcome categories are tabulated on pages 44-
51. In terms of the evidence, ‘abundant’ is coded in green on the tables on pages 21-23, whereas
‘limited’ is coded as red.
In general terms, the evidence base is much larger for issues that are easily measurable; this is
reflected in the abundant evidence for outputs like developing policy documents. However, this
means the evidence base is much more limited on outcomes like affecting political will partly because
it is difficult to measure. This is an underlying issue with the security and justice (S&J) sector in general
and is reflected across all types of literature: researchers tend to concentrate on things that are
measurable.
1 Available on request
June 2019
9
The 2019 update aimed to approach research design more rigorously, particularly with reference to
secondary studies. The intervention, output and outcome categories are provided in Table 1 and
detailed definitions provided in the research protocol (Appendix 1, p.43).
2.3 New and Expanded Areas The Terms of Reference (TOR) outlines seven interventions, outputs or outcomes whose scope is to
be expanded, and one additional category (illicit flows). There was initial testing for some of these
and the searches were adapted to address the areas specified. This is set out below in Table 3.
Table 1: New and expanded research areas
Area Proposed change Additional search terms used
Interventions
Non-state security and justice influencing reform and national level processes
Add in spiritual leaders Spiritual leaders Religious groups Faith groups Religious leaders
Community-based approaches
Add in spiritual leaders Separating the ‘community’ search string into one based on ‘community-based approaches’ and gender. For long-form, the two strings were:
1) Community OR
community-based
approaches OR
community policing
OR community-based
policing OR
empowerment OR
participation OR
grassroots
2) gender-based
approaches OR
women’s groups OR
violence against
women and girls OR
VAWG OR gender
based violence OR
GBV
In the short form, the map kept the current ‘community’ term and added a ‘gender’ term
Spiritual leaders Religious groups Faith groups Religious leaders
Gendered specific interventions and approaches to reform
Expand to VAWG and GBV Using VAWG and GBV as base search terms widened the search considerably. This was
June 2019
10
incorporated to some extent, but much like ‘transitional justice’ the amount of literature would justify its own search.
Outputs
Accountability, effectiveness and transparency
Slight expansion (unspecified) Added to existing search terms and search strings.
Outcomes
Provision is responsive to citizens’ needs
Gender expansion Access to service provision by gender Women’s services Exclusion and access For gender, we had a gender category on the coding form to record whether outcomes/effects in studies have been differentiated by gender.
Equal access to provision Slight expansion (unspecified) Expansion of search terms to include equity.
Security and justice actors are a source of protection, not insecurity
Expansion to cover grievances and extremist narratives
The map added in extremist narratives as an addition to a base search term. We changed base searches to incorporate security AND oversight, for example.
Reduction in illicit financial flows
New category: Indicators can include total value of inward and outward illicit financial flows, or proportion of seized small arms and light weapons that are recorded and traced, in accordance with international standards and legal instruments
Wire/bank transfers Cash carried across borders Import/export of goods and services Transfers of financial assets Transfer of property ownership Cryptocurrencies Tax evasion Proceeds of crime Money laundering Proceeds of corruption Financing of terrorism and conflict Evading capital controls Sanctions busting Bribery
Displacement and refugee-related justice in post-conflict or fragile contexts
New category This was incorporated across a number of searches and added to base search criteria.
June 2019
11
Commercial Justice: Evidence on commercial justice reform, or judicial reform focussed towards providing an enabling environment for economic activity, such as commercial training, commercial courts, contractual obligations, will not be included. Economic issues such as implementation of land or labour rights might be included as part of the evidence on legal empowerment or access to justice.
New category or expansion This was included under the non-state approaches to justice and intervention. Land and labour rights are a big subject but the mapping exercise looked for this in relation to existing search terms.
2.4 Inclusion criteria The review covers two different timeframes -from 2015 onwards for those search terms that were
already included in previous map and from 2008 onwards for the new search terms, as specified in
the Protocol (Appendix 1). This is a pre-defined list of S&J interventions and relevant outcomes agreed
as part of the process. Documents were selected for inclusion according to the following criteria (see
further details in Annex A):
• Relevance: Studies that include information on both S&J interventions and outputs, or
outcomes, irrespective of the nature of the relationship.
• Types of publication: Academic journals, peer-reviewed materials, working papers, grey
literature, and book chapters that were available online at no cost to the reader.
• Date of publication: Materials published from 2008 onwards were included.
• Geographic focus: Low- and middle-income countries.
• Language: Only studies available in English were included.
• Research design: Primary empirical research, evaluation (quantitative or qualitative), and
secondary reviews were included. Theoretical, thematic and conceptual literature was not
considered.
Table 2: Coding Framework
Table 1: Coding framework – intervention, output and outcome categories Intervention categories • Developing accountability mechanisms (at national, regional and local/community level) • Non-state security and justice influencing reform and national level processes • Community-based approaches • Capacity building of organisations • Strategic/statutory frameworks and legislation • Restructuring the security and justice sector • Gender-specific interventions and approaches to reform • Investment in infrastructure and equipment • Integrated political engagement/Activities that promote political will • Demobilisation, disarmament and reintegration • Non-state justice forum and justice facilitator capacity building
June 2019
12
• Legal services • Preventative interventions • Displacement related justice Output categories • Accountability
• Effectiveness and transparency • Non-state actor inclusion in reform processes or negotiations • Community participation and voice • Capacity of state and non-state organisations to deliver • Confidence, trust or satisfaction on the part of citizens in providers • Gender sensitivity and balancing • Roles, coordination and dialogue amongst organisations and agencies • Strategic frameworks developed • Application, compliance and interpretation of laws • State/non-state linkages Outcome categories • Security and justice actors have incentives for improved service delivery • Provision is responsive to citizens’ needs • Resource allocation / funding stability and sustainability • Political will to enact reforms • Ownership of reforms by national and local stakeholders • Citizens’ perceptions of safety and security • Access to provision • Stability and outbreaks of conflict or violence • Security and justice actors are a source of protection not insecurity • Local or national economic development • Judicial redress to protect rights • Human rights measures implemented to improve compliance • Actual crime rates • Legal awareness and confidence • Gender-based violence rates • Poverty reduction • Access to land, inheritance and property rights • Access to public services and economic resources • Women’s empowerment and gender equality
3. Methodology
3.1 Search strategy
The UK Department for International Development’s (DFID) draft theory of change (ToC) for SSR
provided the initial framework of intervention, output and outcome categories for the SSR mapping2.
The search strategy was based on this framework. It utilised a combination of base search terms
relating to various thematic areas of SSR and search specifiers relating to the ToC’s intervention
categories. The S&J search strategy built upon this framework, but replaced the base search terms
with a new set of justice-related themes.
2 Available on request.
June 2019
13
This mapping is not a full systematic review, and therefore cannot claim to capture all relevant
research published in this area. However, the researchers applied a search strategy designed to
capture peer-reviewed materials, working papers, grey literature (e.g. evaluations of donor-led
interventions) and relevant meta-reviews. For each source examined, we recorded the number of
initial search results and the number of these documents that met the inclusion criteria and are
therefore included in the review. The project team searched the same sources that were searched in
the 2015 map, except for one centre which has since closed and two that DFID has requested to be
added to the list. The project also used literature within requested guidebooks, such as the
Stabilisation Unit guidebook.
The scoping report was used to pilot search specifiers to identify the appropriate literature. The
updated search specifiers reflect the increased emphasis within the project on a number of thematic
areas. Two strategies were used, depending on the search capabilities offered by each data source.
For research centres, independent resource centres and similar websites that offer free-text search
capabilities, the project used a defined set of terms.3 Where independent resource centres have
organised their document libraries into pre-defined categories, the relevant key terms or the closest
synonyms to the column 1 search terms were selected. On some research institute websites it is not
possible to search using free-text search terms, and in such cases, relevant documents listed on their
security and justice publications pages were included. For databases like journal indexes that have the
capability to conduct Boolean searches (using logical operators ‘and’ and ‘or’), there was also a set list
of terms.
3.2 Screening
The important element of the screening process is that it consists of two stages. First, the initial
screening was carried out during the searches. The researchers based their judgement on the title
and abstract, incorporating selected articles into a shared Zotero bibliography. The second screening
used a checklist to ensure that items within the shared bibliography were relevant to the map goals.
Further details on the checklist are available in the appendix. The researchers selected materials to be
included in the spreadsheet according to the following criteria:
• Date of publication: For the topics that were covered in the 2015 map, we will restrict the
search to materials produced from 2015 onwards. For new topics, or topics where the
definition of the topic has changed, materials published from 2008 onwards will be included.
• Types of publication: Peer-reviewed journal articles and grey literature, working papers,
evidence synthesis (including systematic reviews, rapid evidence reviews and meta-reviews)
and edited book chapters that are available online at no cost to the reader. Book chapters will
only be included where the text is available electronically directly from the publisher in PDF
full text format. This excludes scanned copies and Google Book previews. Only materials
whose primary purpose is to present empirical evidence will be included. Theoretical studies,
policy statements, guidance notes, and advocacy-oriented materials will not be included.
• Relevance: Studies must explore the relationship between S&J interventions and a given set
of outcomes, irrespective of the nature of the relationship (e.g. positive / negative / neutral).
Donor, international, national government-led, and NGO/CSO level interventions will all be
considered.
3 The Security and Justice Evidence Map Updated Project Protocol (Appendix 1) includes the lists of all terms used within the methodology.
June 2019
14
• Geographic focus: Low- and middle-income countries.
• Language: Only studies available in English will be included.
• Research design: Primary, empirical research or evaluation (quantitative or qualitative) or
secondary evidence synthesis (systematic reviews, rapid evidence reviews/assessments or
meta-reviews). Secondary literature reviews that do not provide a clear methodology will be
excluded. Purely theoretical and conceptual papers will also be excluded.
The following types of materials will be excluded from the catalogue:
• Juvenile and child justice
• War on drugs and drugs regulation
• Environmental justice in fragile contexts
• Probation and parole, except when specifically related to penal reform interventions
• EU’s ‘Area of freedom, security and justice’
• Counter-terrorism and extra-judicial cooperation and interventions
3.3 Coding
Flexibility was built into the coding for the mapping exercises. New intervention, output and
outcome categories identified during the coding process were added progressively, and then back-
coded. Additions and amendments to the coding categories were made during the S&J mapping to
accommodate the justice-orientated literature. The documents selected during the SSR mapping
exercise were recoded using the new S&J categories before the two databases were combined.
Specific detail is provided in the project Protocol document.
Literature that is included after the completion of both screening processes will be coded using the
methodology from the original evidence map. The coding will be based on the intervention,
outcome and output definitions outlined in the appendix. Where relevant, new interventions,
outcomes and outputs can be added throughout the project where agreed by the senior researcher.
3.4 Quality control
The coding was reviewed to ensure quality control. The peer review is also part of the process of
defining and finalising new intervention, output and outcome categories. The review did not include
existing documentation from the previous SSR evidence mapping exercise. There were two review
stages: a sample of the first 40 tranche of documents were checked by lead researchers; a sample of
documents from the full text stage were checked; a sample of the coded documents were checked;
and any code reviews were checked and discussed with the project leads. Throughout, the project
leads were involved in discussions and on hand to provide support to the team.
4. Search results The updated S&J searches generated 2,632,674 articles, of which 85,188 were checked and 550
chosen for inclusion in the first screening. After duplicates were removed, the second screening then
looked at 424 articles and 139 articles were then coded (including 16 duplicates from the 2015 map
that were recoded). This is an inclusion rate of around 25%. For comparison, the 2015 mapping had
an overall inclusion rate of 26% but covered two separate areas of literature. The very specific Security
Sector Reform (SSR) literature had an inclusion rate of 39%, whereas the broader Security and Justice
literature had an inclusion rate of 21%. We consider the inclusion rate of 25% to be quite high and in
June 2019
15
line with what we would expect. The 2015 study had already identified 386 studies, which means that
overall there are now 502 studies in the updated database.
Note that initial systematic searches of large databases tend to generate many results, of which only
a portion are relevant due to not meeting the thematic, methodological or relevance criteria for
inclusion. There is also a lot of generic and anecdotal or purely conceptual S&J literature that is not
relevant to either outputs or outcomes and so falls outside the remit of this map. In addition, once
these studies are excluded and then the initial screening criteria are applied this provides a much more
focussed sub-set than the initial sample.
Reasons for the rejection rate remain the same for this updated map: broad search terms; the role of
indices; duplicate results; and lack of relevance. The use of terms like ‘justice’ generate enormous
numbers of hits that do not meet any of the criteria for inclusion. This is shown graphically below.
Figure 1: PRISMA Diagram
Searches and expert Consultations
Title and abstract Screening
Full text screening
Coding and quality Appraisal
2,632,674 documents (from databases
and websites and expert
recommendations)
85,188 documents screened after the
removal of duplicates
424 documents screened at full text level
139 documents in the final evidence gap map
June 2019
16
5. Features of the evidence base
Figure 2: Research design, publication form and geographic scope of the evidence base
5.1 Research design Figure 2 provides data on the research design, publication form and geographic scope of the evidence
base. 94 per cent of studies use observational research designs. These mainly combine a variety of
qualitative techniques, including interviews, focus group discussion, ethnography, case map analysis,
historical analysis and political economy analysis. Few use quantitative data collection and data
analysis techniques.
The update included only twelve experimental studies and this seems to be a gap, although this may
also reflect the nature of the subject area. The increase in the number of experimental pieces were
largely around police efficiency (e.g. gender sensitivity or high-density patrolling). We are sure that
this is not just a result of the expanded scope but of additional studies.
There are few studies of this type in S&J partly because so much work in this area involves
interventions taking place in complex and politically sensitive environments, where the accessibility
of accurate data may hamper data-based research. However, examples can be drawn from other fields
like social protection where rigorous impact evaluations are being undertaken to explore the impacts
of cash transfers on transient communities in humanitarian and emergency contexts. This shows that
such research is possible if built in to the intervention design itself from inception. It may reflect the
lack of creative thinking in how S&J interventions can be evaluated in the absence of conditions where
‘gold standard’ evaluations can be implemented.
Twenty eight secondary studies were also identified that as in 2015 largely consist of non-systematic
literature reviews and annotated bibliographies. There are very few (2) systematic reviews. The most
significant addition is Denney & Valters (2015) Security sector reform and organisational capacity
15
470
28
0 100 200 300 400 500
Experimental
Observational
Secondary
230
37
58
3
176
Peer-reviewed journal
Book or book chapter
Evaluation
Workshop or conference report
Other report
344
86
13
60
6
Academic organisation/think-tank
Multi-lateral, bi-lateral or inter-…
National/host government organisation
Non-governmental organisation
Private/commercial organisation
94
408
Multi-country
Single country
June 2019
17
building, which is partly based on work related to the 2015 EGM and that focusses on the relationship
between organisational capacity building and accountability, responsiveness and capacity to deliver
security.
5.2 Publication form The largest single group of documents were peer-reviewed journal articles (46%) and ‘other’ reports
(35%). More evaluations (12% rather than the 6% in 2015) met the inclusion criteria in the 2019 map.
The nature of the publications has changed little since 2015. Studies classified as ‘other reports’
include analytical reports and case studies, the majority of which were published by academic
organisations and think-tanks (such as the United States Institute of Peace and the Overseas
Development Institute), followed by non-governmental organisations (such as Saferworld, the Geneva
Centre for the Democratic Control of Armed Forces, and the Open Society Justice Institute). This also
includes reports by multilateral organisations (such as the World Bank, the OECD-DAC and the
International Development Law Organization), and national government organisations (such as the UK
Government Office for Science and the American Bar Association).
Relatively few evaluations have been included. Many of those rejected were evaluations of broader
peacekeeping or fragile states programming, with little explicit detail on S&J. In other cases,
evaluations explored very short timeframes (such as quarterly reports) or commented on progress
against process and management indicators rather than actual results and impacts. More evaluations
were included though relative to the 2015 map. These were mainly mid-term evaluations or final
evaluations, mainly from USAID/DFID.
5.3 Publisher This map covered a marginally lower percentage of academic journals (69% from 72% in 2015). 17%
of material was from multi-/inter-/bi-lateral government organisations (i.e. higher number of
evaluations/reports). The twelve per cent from non-governmental organisations suggests that the
sample has slightly shifted away from academic studies and towards grey material, although some
people may, of course, write in both camps.
5.4 Geographic scope Figure 3 shows that Western Africa remains by far the most studied region. South Asia and East Africa
are the next most studied regions. There are very few studies on Central Asia, East Asia, Eastern
Europe or Northern Africa. It was necessary to use a standard means of classifying the geographic
focus of documents. The UN classification system has been used for this mapping process. See
http://millenniumindicators.un.org/unsd/methods/m49/m49regin.htm for further details. It should
also be noted that the European Union’s ‘area of freedom, security and justice’ was excluded from the
search which ruled out some (a small number) of studies on Eastern Europe, mainly focussed on the
Balkans.
June 2019
18
Figure 3: Number of studies by geographic region
5.5 The quality of evidence Issues of the quality of evidence have not really changed since 2015 in that it remains relatively high
although it should be noted that a critical appraisal of all of the literature is beyond the scope of this
report so this cannot be entirely rigorous. . A good indication of quality is the fact that the majority of
documents in the evidence base are peer-reviewed journal papers and so have been through a
rigorous process of research review. Furthermore, many studies classified as ‘other reports’ have been
published by academic institutions and non-governmental organisations that are renowned for
publishing quality output. Prominent examples include Saferworld, the United States Institute of
Peace, Namati, and the International Development Law Organization, amongst others.
A few features may count against the rigour of evidence. There are few experimental or quasi-
experimental studies, and therefore limited coverage on the presence and nature of causal linkages
between interventions and outcomes, although this may be counteracted by detailed historical
studies looking at political change. Several political-historical studies around cases specifically use a
process tracing methodology that is explicitly seeking to move beyond a historical narrative and to
analyse political causality in decisions-making, i.e. why did people take specific decisions at specific
times? In addition, single country studies are far more prominent than multi-country comparative
studies. Whilst multi-country studies can provide significant insight, there are a number of high quality
case studies that provide a rich basis for the analysis of interventions and results at country level that
can aid understanding of mechanisms in specific contexts.
6. Evidence gap map
Figures 4 and 5 are the evidence gap map. It is presented in two parts for ease of viewing: Figure 4
maps the interventions against outputs, and Figure 5 maps the interventions against outcomes.
36
77
34
28
10
90
18
30
36
3
6
67
37
18
10
36
19
0 100
Global
Eastern Africa
Middle Africa
Southern Africa
Northern Africa
Western Africa
Caribbean
Central America
South America
Central Asia
East Asia
South Asia
South-East Asia
West Asia
East Europe
South Europe
Oceania
June 2019
19
Individual cells show the number of studies that address each combination of interventions and
outputs or outcomes. Green cells show where the review found a significant number of studies –
darker shades of green indicate more studies found. Red cells highlight evidence gaps – darker shades
of red indicate greater evidence gaps.
The gap map only illustrates where there is evidence. It does not provide information on what the
evidence says, nor does it comment on the nature of linkages between interventions and outputs or
outcomes. In particular, dark green cells do not indicate that there is a lot of evidence supporting a
positive relationship between the intervention and output or outcome. Indeed, the evidence may
show positive, neutral or negative effects or be inconclusive. Individual studies may appear in
multiple cells because the majority explore multiple types of intervention, outputs and outcome.
6.1 Overview
Generally, as in 2015, the areas with large evidence bases and the evidence gaps appear where one
would expect them. The gaps appear where the outputs and outcomes are less tangible, and more
difficult to operationalise and measure. As a result, there is less coverage in the literature and less
evidence that interventions make an impact in these areas. The evidence on longer-term
development outcomes such as economic development, poverty reduction, and women’s
empowerment and gender equality is particularly limited. This is also reflected in the secondary
research that tends to draw on those areas where there is primary research. Thematic focus is
shown in figure 4.
Figure 4: Number of studies by thematic focus
Evidence on outputs and outcomes is generally more abundant in combination with those
intervention types that appear more frequently (i.e. ‘capacity building of organisations’,
‘strategic/statutory frameworks and legislation’, ‘community-based approaches’ and ‘restructuring
the security and justice sector’).
92
208
34
11
56
51
134
152
185
43
103
75
9
8
22
0 100 200 300
Defence
Police
Intelligence
Border security
Security and justice governance
Multi-sectoral focus
Non-state actors
Justice sector reform
Access to Justice/Legal empowerment
Prisons
Legal reform
Security promotion/Violence reduction
Urbanisation
Counter-terrorism
Organised crime
June 2019
20
As stated earlier, coding categories were added during the S&J coding process, and then back-coded.
Evidence on many of the newer categories is sparse. Many of these new categories are more justice-
specific, whilst many of the original outcomes and outputs are more general and applicable across a
wider range of intervention types. A full set of definitions is provided as an appendix to the Protocol,
which is on page 43 of this document.
June 2019
21
Figure 5: Evidence gap map – interventions against outputs
June 2019
22
Figure 6: Evidence gap map – interventions against outcomes (in two parts)
June 2019
23
Figure 7: Evidence gap map – interventions against outcomes (in two parts)
June 2019
24
6.2 Analysis: interventions Figure 8: Number of studies by intervention type
Figure 8 displays the number of studies by intervention type. Evidence is abundant for the following
interventions, and many studies provide examples in which a combination of these interventions is
used:
• Non-state actor interventions
• Capacity building of organisations
• Strategic/statutory frameworks and legislation
• Community-based approaches
• Restructuring of the security and justice sector.
Evidence is scarce for the following intervention types:
• Preventative interventions: Few studies were identified that focus on the themes of security
promotion and violence reduction and few include evidence on integrated efforts to prevent
violence and crime, particularly in urban areas. This is not to say that there are few studies on
crime and violence reduction in general. Rather, this means that there are few documents
that our search strategy identified as being specifically and explicitly related to S&J.
• Integrated political engagement/activities that promote political will: This intervention was
not covered significantly in the literature, perhaps because it is not something that is easily
reported on or measured. Studies generally do not discuss these types of activities, even if
they might be happening in the background of other more tangible activities.
• There was a lack of evidence regarding migration and displacement and security and access
to justice. This is a clear area that requires more research.
• The evidence is no longer scarce on non-state justice forum and justice facilitator capacity
building. This reflects an increase in the literature base on interventions aimed at traditional
justice providers and, most frequently, the providers of paralegal services.
• Within the ‘access to justice/legal empowerment’ theme most of the existing evidence is for
the former and there is very little evidence on actual legal empowerment.
167
201
240
288
240
214
161
135
73
75
162
89
13
47
0 100 200 300 400
Accountability interventions
Non-state actor interventions
Community-based
Capacity building
Strategic/statutory frameworks
Restructuring the security and justice sector
Gender specific interventions
Infrastructure and equipment
Political engagement/Political will
DDR
Legal services
Non-state/Justice facilitator capacity building
Displacement and refugee-related justice
Preventative interventions
June 2019
25
6.3 Analysis: outputs Figure 9: Number of studies per output
Figure 9 shows that evidence is abundant for:
• Accountability
• Capacity of state and non-state institutions
• Confidence, trust or satisfaction on the part of citizens in providers
• Roles and co-ordination
• Strategic frameworks
• Community participation
Accountability, the capacity of state and non-state institutions and confidence have all remained
strong areas. In addition, roles and coordination and strategic frameworks remain from the 2015 map.
The only (minor) change here is the inclusion of more discussion regarding the capacity of non-state
institutions.
The literature on community participation in ongoing security & justice initiatives has increased to
abundant since the 2015 EGM. Community participation is often seen in conjunction with
interventions that include ‘community-based’, ‘capacity-building’ and ‘strategic frameworks’. At the
same time, it may also be found in conjunction with outputs that include the terms ‘accountability’,
‘non-state inclusion’, ‘capacity’ and ‘confidence/trust’. This is typically demonstrated through links
between local communities, and their leaders, and security institutions (such as through community
policing initiatives) or initiatives supporting marginalised groups within communities
Figure 9 shows that evidence is scarce for the following outputs, which has not changed since 2015:
• Application, compliance and interpretation of laws
• State/non-state linkages
Application, compliance and interpretation of laws: This is a justice-specific output. Evidence appears
most frequently in combination with the intervention ‘strategic/statutory frameworks and legislation’
followed by ‘capacity building or organisations’ and ‘legal services’. Studies commented on whether
the adoption of new codes, norms and standards had resulted in the fair and consistent application
and interpretation of law by judges and traditional or customary justice facilitators. Evidence on this
267
148
199
326
228
138
187
196
77
95
0 100 200 300 400
Accountability
Non-state actor inclusion
Community participation
Capacity
Confidence/trust/satisfaction
Gender sensitivity/balancing
Roles/coordination/dialogue
Strategic frameworks
Application/compliance/interpretation
State/non-state linkages
June 2019
26
output was often demonstrated through subjective user perceptions or external analysis, and whether
this is considered to be consistent and fair across different types of cases and for different types of
users.
State/non-state linkages: This output is specific to access to justice and legal empowerment. The
evidence is strongest in combination with the interventions ‘community-based approaches’ and
‘strategic/statutory frameworks and legislation’. This output was often expressed in terms of the level
of coherence and linkages between official and customary justice through the recognition, regulation,
institutionalisation or integration of non-state legal orders. This includes codification and the diversion
of cases from state to non-state forums and vice-versa.
6.4 Analysis: outcomes Figure 10: Number of studies per outcome
Evidence is abundant for:
• Political will to enact reforms: Evidence was mostly provided in the form of: opinion or
perceptions on the political will amongst local and national stakeholders to enact reforms or
support interventions; the perceived presence of motivation, commitment or consensus
amongst stakeholders for change; and the presence or absence of spoilers. One would expect
the strongest evidence to appear in combination with ‘integrated political
engagement/activities that promote political will’. This perhaps reflects the fact that there is
little explicit coverage on the intervention ‘integrated political engagement/activities that
promote political will’. There is more coverage on whether other types of activities have
impacted on ‘political will’ as an outcome. Evidence on political will appears more frequently
in combination with the intervention types ‘capacity building for organisations’,
‘strategic/statutory frameworks and legislation’ and ‘restructuring the security and justice
sector’.
• Ownership of reforms by national and local stakeholders: Ownership is a core tenet of S&J,
and one would expect substantial evidence. The evidence base explores this outcome in terms
46
164
185
175
96
179
147
181
109
54
178
62
100
48
13
10
38
39
96
0 100 200
Incentives
Responsiveness
Political will
Ownership
Citizens feel safe and secure
Access to provision
Stability/conflict/violence
Resources/sustainability
Security and justice actors protect
Judicial redress
Human rights
Actual crime rates
Legal awareness/confidence
Gender-based violence rates
Economic development
Poverty reduction
Access to land, inheritance and…
Access to public services and economic…
Women's empowerment and gender…
June 2019
27
of: the extent of perceived ownership by national and local stakeholders; the extent to which
reforms are internally initiated by national and local stakeholders; the involvement of a broad
range of stakeholders (not just elites) in reform; and the extent to which donors take a
controlling role in the planning and implementation of reforms.
• Resource allocation/funding stability and sustainability: The evidence base is particularly
concerned with issues of sustainability after the implementation of interventions. This
outcome was often explored in terms of the absence of resource shortfalls and the extent of
dependency on donor funding.
• Access to provision: Evidence is most likely to be found in combination with the intervention
types ‘community-based approaches’ and ‘legal services’, both of which are primarily
concerned with local-level delivery and engaging citizens where provision might normally be
lacking. Evidence is presented in terms of: the extent of provision across territories; the extent
of access in rural areas; the impact of cultural, financial or gender-related issues upon access;
and the extent to which legal services have enabled access to formal provision.
• Stability and outbreaks of conflict or violence: There is significant evidence on the impact of
interventions on stability, conflict and violence. This output is often discussed in terms of:
political stability or conflict between competing factions and actors; the ability of security and
justice agencies to prevent outbreaks of conflict and violence amongst society; and reductions
in local-level legal conflicts between disputing parties.
• Human rights measures implemented to improve compliance: Human rights are a core tenet
of S&J. The relative strength of evidence for this outcome is perhaps explained by the relative
tangibility of indicators such as: the adoption of relevant legislation, ethical codes or codes of
conduct; human rights training; and vetting and certification and removal procedures to
ensure that personnel are human rights abiding.
• Responsiveness is a new area of abundance since 2015, often in conjunction with
interventions including ‘community-based’, ‘capacity building’, ‘strategic statutory’ and often
focused on in response to gender-based interventions. This is typically demonstrated through
increased representation in security institutions, gender responsiveness in police/justice
institutions, and improved service delivery & customer orientation (e.g. a reduction in
corruption).
Figure 8 shows that evidence is scarce for the following outcomes:
• Incentives for improved service delivery amongst security and justice actors
• Actual crime rates
• Legal awareness and confidence
• Gender-based violence rates
• Judicial redress to protect rights
• Economic development (local or national)
• Poverty reduction
• Access to land, inheritance and property rights
• Access to public services and economic resources
Incentives remains very difficult to identify in the literature and this is reinforced by a general lack of
data across security and justice in general. In particular there is a shortage of data on actual crime
rates and most of what does exist is not clearly correlated with areas of policing. Legal awareness and
confidence is also scarce, but can be explained by the fact that it is quite specific to access to justice
and legal empowerment. Evidence for this outcome appears most frequently in combination with the
June 2019
28
intervention types ‘legal services’ and ‘community-based approaches’. There was no change in the
other categories except for gender-related crime.
Gender-related crime has been removed from the weak category. More studies in this map focussed
on women’s empowerment (largely through increased access to justice), but there is still not enough
robust evidence on the tangible benefits of that empowerment. For example, the evidence base is
limited on the longer term access to economic/public resources or a reduction in GBV rates that
require social norms to change.
7. What has changed?
The updated map has expanded the scope to address some of the gaps identified in the 2015 map,
although the overall findings remain similar. It should be noted that this is also not a simple update,
but the search terms were also changed and expanded, so the differences are not directly
comparable.
7.1 Areas of significant change
• The literature on ‘non-state security and justice influencing reform and national level
processes’ has increased significantly. However, one might still expect more evidence on the
intervention in combination with the outcomes ‘citizens feel safe and secure’ and ‘security
actors protect’;
• The evidence for ‘integrated political engagement/activities that promote political will’
remains much more limited than for the other pre-determined intervention types, even
though it is acknowledged more often. Studies generally do not discuss these types of
activities, even if they might be happening in the background of other more tangible
interventions. There is limited evidence that political engagement/political will interventions
can affect the outcome ‘incentives for improved service delivery’, even though these would
appear to be directly related. Another anomaly is that one might expect more evidence to
appear in combination with the outcome ‘political will to enact reforms’.
• The 2015 map found that there was limited evidence for some types of intervention where
one might expect the evidence base to be larger. In 2015 this included available evidence on
gender-specific interventions. The 2019 map finds that this has improved considerably and is
no longer an absolute gap. There remains limited good quality data related to gender-based
violence rates, however, and it is frequently the case that literature is unable to disaggregate
gender-related data on violence, for example.
• Non-state actors is another area where the 2015 map found scarce evidence but there has
been significant change. The evidence in this area has significantly increased and is no longer
a gap overall. However, the evidence is still relatively weak on the outcome of access to
land, inheritance and property rights. There is also an evidence gap in terms of the political
linkages between on-state and state approaches to S&J.
7.2 Areas related to expansion of search terms
The expansion of the searches also led to some additional areas that are a change from 2015 but
perhaps more difficult to map because it is not clear how far this is a change in the literature or
in the search terms themselves:
• The evidence is not as strong for the ‘developing accountability’ intervention, but there is
abundant evidence on the accountability output. One might have expected the quantity of
June 2019
29
evidence for these directly related interventions and outputs to be similar. Studies comment
on the impact of different types of interventions on accountability as a normative measure
of success, even when interventions to develop accountability are not explicitly mentioned.
• Evidence on preventative interventions remains limited across all categories. One might
have expected more evidence for the outcomes: stability and outbreaks of conflict or
violence; actual crime rates; and gender-based violence rates. This was very low in 2015 and
remains a gap in 2019.
• A significant gap appears related to issues of migration, displacement and refugees. This was
a specific set of search terms and was itself a difference from 2015. It is possible that these
terms are not the right ones for getting at this specific area. The academic literature on
migration, for example, is very well developed, but dislikes terms that securitize refugees. It
is possible that looking for search terms based on security language may not be getting to
some of this literature.
7.3 Detailed changes
Overall the database was updated and expanded by removing 8 articles and updating links for 72
articles as well as adding 123 new articles. Changes were made to make the coding more systematic.
More emphasis on gender, empowerment, displacement and mutual legal assistance was added in
the search process and the map placed additional emphasis on methodological rigour and
transparency.
• This did lead to some changes within the overall data, as one might expect, and a 30% overall
change in documents. However, the increase was uneven. Literature on
interventions/outcomes to do with access to justice or non-state justice show a
disproportionate increase. This is particularly clear in relation to gender.
• East and Middle Africa also showed particularly large increases. West Africa also showed a
large increase, but from a higher starting point, as the most studied region in the 2015 EGM.
West Asia and the Americas/Caribbean also shows significant increases. There were no
additional studies in North Africa or East Asia.
• In terms of research design there was a big increase (88%) in experimental approaches and
also in secondary analysis (87%) but these were from low bases and remain relatively small in
absolute terms. Other increases were seen in smaller categories like evaluation (141%) but
again these are relatively small numbers (24 total). Overall there was also an increase in
publications from governments relative to academic research and a larger increase (33%) in
single case rather than multi country studies (18%).
• In terms of thematic focus, intelligence increases by 70%, border security 83% and S&J
governance 81% but this is misleading for two reasons. First, these increases are from
relatively low levels (20, 6 and 31 respectively). Secondly, and more importantly, these are
largely studied within analyses of wider SSR reforms. Very few, if any, pieces focus on these
exclusively.
• Non-state actors (65%), access to justice/legal empowerment (53%), and security
promotion/violence reduction (108%) have also increased. These are more accurate figures
to use, reflecting a disproportionate increase of literature specifically on these subjects. In the
same way as experimental approaches have increased from a low base, work on urbanisation
(80%) and counter-terrorism (+0%) have also increased significantly but are still relatively
small in absolute terms (5 to 9 and 10 respectively).
June 2019
30
• Geographically, there has been a large increase in some areas: East Africa (51%), Middle Africa
(42%), Southern Africa (33%), Caribbean (50%), Central America (+6%), South America (39%),
West Asia (+4%), and Oceania (+6%). However, some of these are from a low base and there
are still significant gaps in some important areas including North Africa (0%), East Asia (0%),
East Europe (11%), and Southern Europe [Balkans] (13%).
• The interventions that have increased the most have been in non-state subjects (47%), gender
specific (55%), legal services (50%), non-state/justice facilitator (112%) and preventative
interventions (124%), although this has increased from a relatively low level (21 to 47). There
is a significant gap here in work on migration and security and justice.
• The outputs have seen a fairly proportional change across the board with the highest change
in confidence/trust/satisfaction (31%) and the lowest in Strategic frameworks (7%).
• Outcomes are more varied, however but include some very significant increases in: Incentives
(35.3%), responsiveness (31%), ownership (45%), citizens feel safe and secure (35%), access
to provision (41%), resources/sustainability (33%), judicial redress (50%), legal
awareness/confidence (100%), access to land, inheritance and property rights (100%), access
to public resources (39%), and women’s empowerment (146%).
• The number of studies reporting on poverty reduction reduces by 17%, but this may be an
anomaly of the methodology due to the removal of the ‘no intervention’ intervention and the
removal of inaccessible documents.
8. Potential avenues for future research The evidence base for security and justice programming is generally considered to be weak. ‘Much of
the literature is normative, presenting recommendations with little empirical evidence about what
works. There is little in the way of rigorous evaluation on the effects of institutional reform
programmes on security and justice provision’ (Bakrania, 2014: p. 22). Whilst 386 documents in the
original map plus a further 123 in the update is a substantial sample of evidence upon which to base
assessments about evidence gaps, one could argue that our rates of inclusion (and the large number
of documents rejected for coding) reinforce this widely held belief.
This is reinforced by the very few experimental or quasi-experimental studies in the database and thus
analysis on the causal linkages between interventions and outputs or outcomes is lacking. This is
particularly concerning given that little is known on the impact of S&J programming on broader
development outcomes. Observational quantitative studies, for example, could provide evidence on
correlation and complement the largely qualitative methods used in the majority of papers in S&J.
They may also provide useful evidence on impact of interventions, which remains limited.
The analysis suggests that there is very limited evidence on the impact of S&J interventions on broader
development outcomes such as crime rates, gender-based violence rates, economic development and
poverty reduction. The updated map reinforces the findings of 2015 that diversifying the research to
include more impact evaluations based on experimental and quasi-experimental approaches, and
systematic or non-systematic reviews focusing on specific S&J themes and interventions, may help
build a more rigorous evidence base. The Denney & Valters (2015) Security sector reform and
organisational capacity building paper and the Jackson and Bakrania (2018) paper (see bibliography)
based on the 2015 map would reinforce that since they both draw wider conclusions.
The nature of data available, where results are long-term and frequently intangible makes it very
difficult to measure, as does the nature of gathering data in difficult and unstable environments.
However, there has been an increase in impact evaluations within some of those environments. The
June 2019
31
difficulty in reaching some of these is in getting hold of the grey literature that may exist outside public
databases, e.g. within private companies and sub-contractors.
The following sections highlight areas of abundant evidence that may be suitable for synthesis
research, and areas of limited evidence where further impact evaluation may be desirable.
8.1 Areas suitable for synthesis research There are a relatively large number of studies in the thematic areas and these remain relatively
stable in the 2019 map:
• Police
• Defence
• Access to justice/legal empowerment
• Justice sector reform.
There are also a relatively large number of studies for the intervention areas:
• Capacity building of organisations
• Strategic/statutory frameworks and legislation
• Community-based approaches
• Restructuring the security and justice sector.
• Non-state actor interventions
Since there has been a relatively limited number of studies in these areas since 2015, this area remains
a useful one to consider for synthesis reports, particularly since so many comprise case studies.
8.2 Priority research areas The research process for the 2019 update was broader than for 2015 and included a scoping report
that included discussions around priority areas. It also reflects the changes outlined above. However,
the updated map does still show a number of areas where there could be a primary research priority:
• Developing accountability mechanisms (at national and local/community level): the evidence
is abundant on the accountability output, but remains not as strong on the ‘developing
accountability’ intervention. Studies still seem to comment on the impact of different types
of interventions on accountability as a normative measure of success, even when
interventions to develop accountability are not explicitly mentioned.
• Non-state security and justice influencing reform and national level processes: the strength
of evidence for combinations with the outcomes ‘citizens feel safe and secure’ and ‘security
actors protect’ is limited. With the increase in work on non-state actors one might have
expected more evidence for these combinations but there seems to be a lack of linkages
between non-state actors and other mechanisms.
• Integrated political engagement/activities that promote political will. This is a very difficult
area to measure and there are key areas where there is a lack of data even when normative
claims are frequently made, like a political dividend in the form of legitimacy in return for
service provision. ‘Political will’ is also something that lacks strong evidence and is either seen
as a form of spoiler to external reform, or as something that is impacted as an outcome.
• Displacement and refugee-related justice: This is clearly a key area for contemporary S&J
interventions but there is a clear lack of literature and evidence in this area that cuts across
different approaches and links outputs to outcomes. This is a clear area of need.
• There were no studies that focused on mutual legal assistance or direct bilateral judicial
cooperation. The only studies that focused on this that were found during the search process
June 2019
32
were introductory or conceptual studies, or did not focus on a low- or middle-income country
as a recipient.
The 2015 map also identified a number of areas where evidence was limited and the 2019 map finds
some of these sub-areas remain relatively short of evidence:
• The quantifiable impact of gender-specific interventions, specifically the availability of gender-
based data, including gender-based violence rates. However, the amount of gender-related
evidence has increased since 2015.
• Infrastructure and equipment.
• Demobilisation, disarmament and reintegration: perhaps quite surprising but most evidence
is descriptive and technical, and does not address important issues like long-term effects of
demobilisation.
• For non-state justice forum and justice facilitator capacity building, the evidence has increased
since 2015 but remains limited in some outcome categories. Land rights remain an issue for
evidence, as does the ability of non-state actors to influence reform structures.
• Legal services and para-legal support.
• Preventative interventions remain limited across all categories.
All of these intervention types may serve as a basis for further impact evaluation.
June 2019
33
References A searchable list of the documents included in the rigorous review is available in the accompanying
database: https://gsdrc.org/wp-content/uploads/2019/06/SJEvidenceMappingDatabase2019.xlsm.
Asian Development Bank. (2009). Legal empowerment for women and disadvantaged groups.
Manila: Asian Development Bank
Bakrania, S. (2014). Safety, security and justice: Topic guide. Birmingham: GSDRC.
Banerjee, A., Chattopadhyay, R., Duflo, E., Keniston, D., & Singh, N. (2012a). Improving police
performance in Rajasthan, India: Experimental evidence on incentives, managerial autonomy and
training. New Delhi: National Bureau of Economic Research
Banerjee, A., Chattopadhyay, R., Duflo, E., Keniston, D., & Singh, N. (2012b). Can institutions be
reformed from within? Evidence from a randomized experiment with the Rajasthan police. New
Delhi: National Bureau of Economic Research
Barendrecht, M., Kokke, M., Gramatikov, M., Porter, R., Frishman, M., & Morales, A. (2013). Impact
assessment of the Facilitadores Judiciales programme in Nicaragua (WIDER Working Paper No.
2013/113). Helsinki: UNU-WIDER
Cox, M. (2008). Security and justice: Measuring the development returns. London: Agulhas
Development Consultants
Cox, M., Duituturaga, E., & Scheye, E. (2012). Building on local strengths: Evaluation of Australian law
and justice assistance. Canberra: AusAID
Denney, L and Valters, C (2015) Evidence Synthesis: Security Sector Reform And Organisational
Capacity Building Rapid Evidence Assessment, November, DFID
European Commission. (2011). Thematic evaluation of European Commission – Support to justice
and security system reform. Brussels: European Commission
García, J.F., Mejía, D., & Ortega, D. (2012). Police reform, training and crime: experimental evidence
from Colombia's plan cuadrantes. Caracas: CAF
Jackson, P and Bakrania, S (2018) Is the Future of SSR non-linear? Journal of Intervention and
Statebuilding, 12:1, 11-30, DOI: 10.1080/17502977.2018.1426548
Maru, V. (2010). Access to justice and legal empowerment: a review of World Bank practice.
Washington D.C.: The World Bank.
OECD-DAC. (2012). Evaluating peacebuilding activities in settings of conflict and fragility: Improving
learning for results. Paris: OECD
Roseveare, C. (2013). Rule of law and international development. London: DFID.
Sandefur, J., & Siddiqi, B. (2013). Delivering justice to the poor: Theory and experimental evidence
from Liberia. Washington, DC: World Bank Workshop on African Political Economy
SAS. (2013). Tools for measurement, monitoring and evaluation: Sources of conflict, crime and
violence data. London: DFID / Small Arms Survey.
Wilson, D. B., Parks, R. B., & Mastrofski, S. D. (2011). The impact of police reform on communities of
Trinidad and Tobago. Journal of Experimental Criminology, 7(4), 375-405
June 2019
34
Appendix 1
Security and Justice Evidence Map Updated Project Protocol
Updated after Scoping Report
Objectives This rigorous evidence mapping exercise will expand upon the existing security and justice evidence
map completed in 20154, bringing it up to date and extending it to cover new subjects of interest.
The ultimate objective is to provide a catalogue of reliable, rigorous evidence that DFID can draw
upon to inform current thinking about how S&J programming can contribute towards reaching
Sustainable Development Goal 16 (‘Peace, Justice, and Strong Institutions’).
This project will build upon the existing catalogue compiled in 2015, and will retain the existing data
within it, by completing the following tasks:
• Check and update web links to all items already catalogued (386 items).
• Adjust and test the search keywords used to identify the literature that will be screened for
inclusion.
• Where the search strings have not changed, the project will identify and catalogue new
materials published for the period 2015-2019.
• Where the search strings have changed, the project will identify and catalogue new
materials published for the period 2008-2019.
The existing catalogue of evidence and visual evidence map, stored in an Excel spreadsheet, will be
updated to incorporate the new material identified through this exercise.
We will maintain compatibility with the methodology used to produce the 2015 evidence map to the
fullest extent possible. It is likely that some of the sources searched in 2015 might no longer exist, or
may have made changes to their search engines and/or indexing methods, so it might not be
possible to exactly replicate the 2015 map, but we will attempt to maintain consistency as much as
possible.
Project Outline
Sources to be searched
This mapping is not a full systematic review, and therefore cannot claim to capture all relevant
research published in this area. However, the researchers will apply a multi-pronged and rigorous
search strategy designed to capture peer-reviewed materials, working papers, grey literature (e.g.
evaluations of donor-led interventions) and relevant meta-reviews. For each source examined, we
will record the number of initial search results and the number of these documents that meet the
inclusion criteria and are therefore included in the review. The project team will search the same
sources that were searched in the 2015 map, except for one centre which has since closed and two
4 Shivit Bakrania, (2015) Security and justice evidence mapping. Governance and Social Development Resource Centre, University of Birmingham, Birmingham, UK
June 2019
35
that DFID has requested to be added to the list. The project will also use literature within requested
guidebooks, such as the Stabilisation Unit guidebook.
Citation indexes and bibliographic databases
• Google Scholar
• JSTOR
• IngentaConnect
Research institutes
• Bingham Centre
• The Danish Institute for International Studies
• The Geneva Centre for the Democratic Control of Armed Forces / ISSAT document library
• Harvard Kennedy Program in Criminal Justice
• The Hague Institute for Innovation of Law (HiiL) Innovating Justice
• INPROL
• International Development Department, University of Birmingham/GSDRC
• The Justice and Security Research Programme (LSE)
• Namati
• Open Society Justice Initiative
• Overseas Development Institute
• United States Institute for Peace
• Vera Institute of Justice
• IDRC Digital Library [on recommendation of expert]
• Centre of Law and Social Transformation [on recommendation of expert]
• IDLO [on recommendation of expert]
• Igarapé [on recommendation of expert]
Other websites
• AusAID/DFAT
• Eldis
• Knowledge Platform – Security and Rule of Law
• OECD-DAC
• Political Settlements Research Programme
• Research For Development (DFID)
June 2019
36
• Saferworld
• SSR Resource Centre [closed]
• Stability Journal
• USAID
• World Bank
Expert Consultation
Consultation with experts will run alongside the online literature searches. Where possible, the same
set of policymakers, practitioners and academics that were consulted for the 2015 map will be
contacted. Other experts will also be identified and added to the expert list. These consultations will
seek updated recommendations for literature. These experts’ involvement will be on a voluntary
basis and we cannot guarantee their participation.
Expert name Institution
Peter Albrecht Danish Institute of International Studies / UNSOM
Tom Carothers Carnegie
Rachel Kleinfeld Carnegie
Freddie Carver SU
Lisa Denney ODI
Deval Desai World Bank
Pilar Domingo ODI
Timothy Donais Balsillie School of International Affairs
Heidi Hudson University of the Free State, Bloemfontein
Robert Muggah IGARAPE
Eric Scheye Independent Consultant
Erwin Van Veen Clingendael
Ilaria Bottigliero IDLO
Peter Chapman Open Society Foundation
Marcus Cox Agulhas
Adrian Di Giovanni IDRC
Deborah Isser World Bank
June 2019
37
Thematic Scope The project covers not just the existing themes of interest but has expanded to cover a number of
additional emerging themes. This section details the scope of the expanded area of research.
Community, Empowerment and Gender
The theme of community has been divided into the two themes of empowerment and gender. The
increased emphasis on gender is designed to increase the amount of literature that focuses on
gender-based interventions including those implemented by women-led organisations and
interventions with a gendered purpose. This reflects the increased emphasis on issues such as
gender-based violence and violence against women and girls. These approaches are already
included under the existing intervention definitions.
Mutual Legal Assistance
The theme of mutual legal assistance is designed to identify literature that focuses exclusively on
international legal cooperation, particularly in the context of transnational organised crime. This has
developed out of the emerging theme of organised crime, which is being covered separately by
DFID. The theme of mutual legal assistance, instead, focuses explicitly on bilateral or multilateral
interventions aimed at improving judicial cooperation. This has been added as an intervention
category. The theme of mutual legal assistance excludes topics such as counter-terrorism and extra-
judicial cooperation, focusing on the capacity building and cooperation efforts instead.
Commercial Justice
The overarching theme of commercial justice is not within the scope of this research project.
However, the sub-theme of land governance and rights is closely related to many access to justice
and legal empowerment initiatives, often with an emphasis on increasing property, land and
inheritance rights for women and other marginalised groups. Increased access to land, inheritance
and property rights is already an outcome, however there was a question as to whether we should
incorporate interventions tailored towards land rights as an intervention in their own right, even if it
was not explored to the same extent as other categories. We argue that this would be inappropriate
for the following two reasons.
First, as land rights interventions are closely related to a number of existing interventions, including
gender, empowerment and legislation, we expect literature on land rights to be identified using the
existing search terms. This is particularly the case given the increased emphasis on gender and
empowerment within the EGM. As detailed below, empowerment and gender will both be studied
in the short-form searches for the period 2008-2019 and gender will be studied in the long-form
searches for 2008-2019. In the initial work on the search protocol in the scoping report, papers
focused on land rights were present in the results, so we would expect the search methodology to
identify an appropriate amount of literature on this subject.
Second, exploring one intervention using a tailored or reduced methodology would undermine the
purpose of the evidence gap map, which is to explore and map the literature on specific thematic
areas. Using multiple approaches within the methodology provides questionable value if the goal is
to compare the evidence between thematic areas.
Displacement
The theme of displacement aims to identify literature that focuses on whether security and justice
interventions influence the ability of internally and externally displaced individuals to access justice
June 2019
38
mechanisms. In particular, this emphasises that displaced individuals are able enjoy their rights; are
not excluded from local and national justice processes; and, in the longer run, are able to access
restitutive processes in both the restoration of housing, land or property and in justice for rights
violations beyond transitional justice. Displacement and refugee-related access to justice has been
added as an intervention. We may find distinct displacement-specific outputs and outcomes during
the research project, which will be explored in the coding stage of research.
Non-Intervention
The discrete intervention of ‘Non-intervention’ has questionable contribution. In the original
evidence map the non-intervention category was recorded just six times and we do not expect to
uncover much further literature that comes under this category. As this evidence map is focused on
the relationship between interventions, outputs and outcomes, we argue that this category can be
removed to further streamline the process of screening and coding. This will involve the removal of
6 items from the original database.
Search Terms The scoping report was used to pilot search specifiers to identify the appropriate literature. The
updated search specifiers reflect the increased emphasis within the project on a number of thematic
areas. The use of search specifiers will be flexible and adaptive in nature to make the most of the
capabilities of each search engine.
Two strategies will be used, depending on the search capabilities offered by each data source. For
research centres, independent resource centres and similar websites that offer free-text search
capabilities, the terms in table 1 will be used. Where independent resource centres have organised
their document libraries into pre-defined categories, the relevant key terms or the closest synonyms
to the column 1 search terms will be selected. On some research institute websites it is not possible
to search using free-text search terms, and in such cases, relevant documents listed on their security
and justice publications pages will be included. For databases like journal indexes that have the
capability to conduct Boolean searches (using logical operators ‘and’ and ‘or’), the terms in table 2
will be used.
Table 1: Search terms and synonyms for databases without Boolean search capability
For databases and websites that provide free-text searching without Boolean (AND/OR) capability,
use each of the terms in column 1 in combination with each of the terms in column 2. This creates 2
x 12 = 24 terms to run on each source, such as “security and justice accountability” and “justice
capacity building”.
There have been two changes to the original proposal. First, Community has been divided into
empowerment and gender. Second, the three emerging themes of displacement and mutual legal
assistance have been added to reflect the increased emphasis on these subjects by DFID. The
original specifier of organised crime has been subsumed into the mutual legal assistance specifier to
reflect the emphasis on legal cooperation.
Column 1: Base
Column 2: Search Specifiers
June 2019
39
Search Terms
“Security and Justice” or “Justice”
Accountability
Non-state actors
Empowerment
Gender
Capacity building
Legislation
Evaluation
Counter-terrorism
Urbanisation/urbanization
Displacement
Mutual legal assistance
Table 2: Search terms and synonyms for databases with Boolean search capability
For databases that provide Boolean (AND/OR) search capabilities, such as journal indexes, use the
term in column 1, ‘AND’ed with each term in column 2 in turn. This produces 6 x 12 = 72 searches to
run on each source, where each search contains two main clauses that are ‘AND’ed together, with
many ‘OR’ clauses to broaden the search.
There have been a number of changes to the original proposal. Non-state actors has been updated
to incorporate faith groups. Community has been split into two separate searches. The first,
community, incorporates the non-gender searches and has added spiritual leaders. The second,
gender, reflects the emphasis on gender-based approaches and gender-based violence. The
emerging themes of mutual legal assistance and displacement have been added. The category of
organised crime has been replaced by the mutual legal assistance category. The emerging theme of
counter-terrorism has been expanded to reflect the emphasis on security forces as a potential
source of animosity.
Column 1: Base Search Term Column 2: Search Specifiers
Security and Justice Justice Judicial Reform Legal Empowerment Access to Justice Penal Reform
Accountability OR grievance mechanisms OR ombudsman OR oversight OR parliamentary committees OR financial management OR complaints commissions OR victim support OR dispute resolution
Non-state actors OR informal OR multi-layered OR customary OR traditional OR civil society OR non governmental organisations OR human rights organisations OR women’s organisations OR religious organisations OR media OR non-state armed groups OR rebel groups OR warlords OR militias OR vigilantes OR criminal groups OR faith groups
June 2019
40
Community OR community policing OR community-based policing OR empowerment OR participation OR grassroots OR spiritual leaders OR tribal OR elders
Women’s services OR gender exclusion OR gender-based approaches OR women’s groups OR violence against women and girls OR gender based violence
Capacity building OR train OR organisational development OR technical OR professionalism OR leadership OR ministry of internal affairs OR ministry of justice OR ministry of finance OR budgets OR human resources OR databases OR [demobilisation/demobilization] OR salaries OR mentoring OR skills OR incentives
Legislation OR regulation OR statutory frameworks OR strategic frameworks OR planning and coordination OR review commission
Evaluation OR monitoring OR indicators OR review OR results OR outcome OR impact
[Emerging theme] Mutual legal assistance OR judicial cooperation OR transmission of information OR letters rogatory OR transfer of proceedings OR recognition of judgement
[Emerging theme] Counter-terrorism OR extremist narrative OR grievances
[Emerging theme] External displacement OR internal displacement OR refugee hosting
[Emerging theme] Urbanisation/urbanization
Based on the experience of the 2015 map, the following search terms were not used, and will not be
used in this project:
• Transitional justice was removed as a base search term, as it is a broad theme in its own
right, and although related to S&J, covers a number of different areas. This decision was
confirmed during the scoping report.
• Commercial justice was considered too broad for the scope of the project. As argued above,
an additional focus on land rights, above what will already be uncovered through the
existing methodology, would also be outside of the scope of the map.
• Legal reform was removed as a search term because it is a large and complex thematic area
and covers many sub-themes not related to S&J.
• 'Legal services' did not generate relevant results.
• ‘Migration’ and ‘Climate Change’ did not generate many relevant results.
June 2019
41
Search Process In order to ensure that the research project is both thorough and delivered on time, the scope of
research is the following. Where there have been no changes, or only minor alterations, to the
original search terms, the map will update the original database for the period 2015 – 2019. Where
existing articles in the database are found during the screening stage of the expanded categories,
these will be included for re-coding. The literature searches will be used with the following time
restrictions:
Type of search Search terms Search period
Short Form Accountability, non-state actors, capacity building, legislation, evaluation, counter-terrorism, urbanisation
2015 - 2019
Empowerment, gender, displacement, mutual legal assistance
2008 - 2019
Long form Accountability*, capacity building, legislation, evaluation, counter-terrorism, urbanisation, non-state actors, community*
2015 – 2019
Gender, mutual legal assistance, displacement
2008 - 2019
Inclusion / Exclusion Criteria The researchers will select materials to be included in the spreadsheet according to the following
criteria:
• Date of publication: For the topics that were covered in the 2015 map, we will restrict the
search to materials produced from 2015 onwards. For new topics, or topics where the
definition of the topic has changed, materials published from 2008 onwards will be included.
• Types of publication: Peer-reviewed journal articles and grey literature, working papers,
evidence synthesis (including systematic reviews, rapid evidence reviews and meta-reviews)
and edited book chapters that are available online at no cost to the reader. Book chapters
will only be included where the text is available electronically directly from the publisher in
PDF full text format. This excludes scanned copies and Google Book previews. Only materials
whose primary purpose is to present empirical evidence will be included. Theoretical
studies, policy statements, guidance notes, and advocacy-oriented materials will not be
included.
• Relevance: Studies must explore the relationship between S&J interventions and a given set
of outcomes, irrespective of the nature of the relationship (e.g. positive / negative / neutral).
Donor, international, national government-led, and NGO/CSO level interventions will all be
considered.
o The relevance criteria will be applied in a two-stage screening process. First, the
initial screening will be carried out during the searches. The researchers will base
June 2019
42
their judgement on the title and abstract, incorporating selected articles into a
shared Zotero bibliography.
o The second screening will use a checklist to ensure that items within the shared
bibliography are relevant to the map goals. Further details on the checklist are
available in the appendix.
• Geographic focus: Low- and middle-income countries.
• Language: Only studies available in English will be included.
• Research design: Primary, empirical research or evaluation (quantitative or qualitative) or
secondary evidence synthesis (systematic reviews, rapid evidence reviews/assessments or
meta-reviews). Secondary literature reviews that do not provide a clear methodology will be
excluded. Purely theoretical and conceptual papers will also be excluded.
The following types of materials will be excluded from the catalogue:
• Juvenile and child justice
• War on drugs and drugs regulation
• Environmental justice in fragile contexts
• Probation and parole, except when specifically related to penal reform interventions
• EU’s ‘Area of freedom, security and justice’
• Counter-terrorism and extra-judicial cooperation and interventions
Data to be Recorded in the Catalogue
• Bibliographic data (authors, publication date, source)
• Hyperlink to the publication
• Abstracts/summaries (NB: Abstracts or summaries will be included where readily available
and exportable into the database. Not all documents will have abstracts or summaries).
• Research characteristics (see table 3 below)
• Intervention, output, and outcome categories as specified in the terms of reference
• Geographic focus by region
• Sub-sector addressed by the intervention (e.g. police / military / non-state actors /
intelligence / border control / holistic or sector-wide / governance/ access to justice / legal
empowerment etc.).
It would be impossible to undertake a thorough quality assessment of all of the materials selected
for inclusion in the database. However, we will record certain characteristics of the studies which
provide some information about quality and relevance, and which can be easily determined.
Database users will be able to use this information to judge the size and quality of the evidence
base.
June 2019
43
Table 3: Research characteristics
Characteristics Possible classifications
Publication form o Peer-reviewed journal article
o Book chapter
o Evaluation
o Workshop or conference report
o Other report
Publisher o Academic organisation or research institute
o Multilateral or inter-governmental organisation
o National government organisation
o Non-governmental organisation
Geographic
scope
o Multi-country comparative analysis
o Single country studies
Research
design5
o Experimental or quasi-experimental
o Observational
o Secondary review (systematic and other literature reviews
with a clear methodology)
Coding Literature that is included after the completion of both screening processes will be coded using the
methodology from the original evidence map. The coding will be based on the intervention,
outcome and output definitions outlined in the appendix. Where relevant, new interventions,
outcomes and outputs can be added throughout the project where agreed by the senior researcher.
Quality Control The coding will be reviewed to ensure quality control. The peer review is also part of the process of
defining and finalising new intervention, output and outcome categories. The review will not include
existing documentation from the previous SSR evidence mapping exercise. There will be two review
stages:
5 Based on DFID How-to Note Assessing the Strength of Evidence, 2014
June 2019
44
1. The first 40 documents screened at title and abstract stage will be checked by the lead
researchers. Any discrepancies or inconsistencies will be discussed, and amendments will be
made.
2. The first 40 documents screened at full text stage will be checked by the lead researchers.
Any discrepancies or inconsistencies will be discussed, and amendments will be made.
3. The first 20 documents coded and entered into the database will be checked by the lead
researchers. Any discrepancies or inconsistencies will be discussed, and amendments will be
made.
Any new intervention/output/outcome codes added during the cataloguing process will be reviewed
and recoded after both review stages.
Evidence Mapping This assessment will produce both a series of graphical outputs and a narrative report commenting
on the evidence trends. Graphical outputs will include:
• An evidence map: this will compare interventions to outputs/outcomes, providing an easily
accessible illustration of the number of studies and their characteristics indicative of quality.
• A series of charts comparing interventions to outputs/outcomes illustrating the quantity of
evidence occurrences, disaggregated by methodology, focus, or region.
The narrative report will provide an analysis of the characteristics of the evidence base, including:
• Commenting on the nature of the evidence base.
• An assessment of the scope of the evidence base – disaggregated by methodology, focus,
region, country interventions, and outcomes/outputs.
• A description of where evidence is abundant, patchy, or missing – referring to the evidence
map and commenting on the characteristics of evidence available for different category
types.
• A clear and accessible summary of the key evidence gaps as a means of highlighting
potential areas for future research.
Personnel
Role and name Duties in the project Relevant experience
Principal investigator: Paul Jackson
Design project and search methodology; guide and train the research assistants; coordinate support from other project team members; monitor and manage the project; review research assistants’ work throughout the project; lead the drafting of the final report.
Professor of African Politics, University of Birmingham. More than twenty years’ experience in high-profile research and policy development. Senior Security and Justice Adviser to the UK Government Stabilisation Unit and adviser to the UK Government’s Head of Profession (Conflict). Significant senior research management and participation roles at the national and international level.
June 2019
45
Research Fellow: Shivit Bakrania
Assist in designing the project and search methodology; train and guide research assistants; assist in drafting the report.
Research consultant and social research methods specialist with 15 years’ experience in international development. Specialised in producing research and evidence tools to support evidence-based policy and programming, including experience producing evidence maps for DFID and Unicef.
Research assistant: Joe Bell
Search through source databases following to identify and select materials for inclusion; classify and catalogue materials correctly; assist in drafting final report.
Joe Bell is a doctoral researcher with a very strong academic background, having received distinctions in his postgraduate degrees and a full ESRC scholarship for his PhD work.
Database/analysis tools & project support: Brian Lucas
Assist in designing the project and search methodology; assist in training research assistants and in quality-checking their work; set up database and data analysis tools.
Project manager specialising in research communication and knowledge management. Twenty-seven years of professional experience including extensive experience working with metadata catalogues and delivering products presenting and communicating information.
June 2019
46
Appendix 1: Intervention, Output and Outcome Category Definitions
Intervention
In line with ToR requirements, studies should be selected that explore the relationship between S&J
interventions and a given set of outcomes, irrespective of the nature of the relationship (e.g. positive
/ negative / no evidence of a relationship found). The coding of outputs and outcomes should follow
this logic.
Intervention Definition
Developing accountability mechanisms (at regional, national and local/community level)
Interventions directed at enhancing and developing accountability mechanisms and democratic oversight, to ensure that security and justice agencies meet expected standards of behavior and performance, including the prevention of abuses and violations by personnel. At the local community-level, this includes interventions to develop the capacity of civil society organisations to actively monitor security and justice policies and practices. This may also include legal empowerment initiatives, which aim to make the law more responsive to the needs of the poor and the vulnerable, and to assist them in protecting their rights and demanding accountable governance. At the national level, this includes interventions to develop and strengthen oversight mechanisms at the executive level. It also includes supporting independent oversight institutions, including the legislature (parliament), the judiciary, and other independent bodies with specialised mandates in the governance and accountability of the security and justice sectors.
Non-state security and justice influencing reform and national level processes
Interventions directed at legal or statutory non-state actors (including spiritual leaders) to support their engagement in the reform process at a national level. This includes supporting the engagement of diverse groups in dialogue on security and justice, supporting confidence building between the security and justice sectors and civil society, or supporting civil-society to undertake outreach and awareness raising programmes. This includes interventions initiated by non-state actors, such as nongovernmental organisations and non-state or informal armed groups, to influence national-level processes. This category also includes interventions directed at extra-legal or non-statutory non-state actors who may influence security and justice processes at a national level in different ways through DDR, integration, transitional justice and peace processes.
Community-based approaches
Interventions directed at local-level developments and actors, including spiritual leaders where appropriate. This includes interventions to support the engagement of marginalised and vulnerable groups, and to promote formal interaction between communities, customary justice mechanisms, local officials and authority, and security and justice agency representatives. This also includes legal empowerment initiatives, which aim to use the law to benefit the disadvantaged, in terms of assisting people to protect their rights, pursuing their economic interests and demanding accountable and responsive governance. This also includes attempts to communicate information about the functioning of security and justice institutions and respective reforms to the wider public.
June 2019
47
Organisational capacity building interventions
Interventions to improve the institutional and organisational capacity of state security and justice institutions and agencies. This will typically be technical assistance comprised of training and organisational development interventions focussed on agencies and the appropriate ministries governing them. Capacity building interventions include those focussing on increasing capabilities, professionalisation and leadership, as well as in areas such as improving human rights compliance, access to redress and gender equality. This includes the development of professional legal education and qualifications.
Strategic/statutory frameworks and legislation
Interventions focussed on national security and justice policies and strategies, and the statutory and legal frameworks to address the security and justice needs of citizens and the state. These laws, standards, policies, strategies, plans and frameworks are typically based on a country’s constitution and legislation. They concern the relationship between security and justice providers and the public, and clarify behaviours, roles and responsibilities of state institutions in providing security and justice. This includes interventions towards the adoption and/or reform of legal codes, norms and standards, with the aim of improving clarity, to specify the legal position of non-state legal orders in relation to official law, and to ensure compliance with international standards of fairness, due process and human rights norms. At another level, this includes efforts to provide strategic national guidance to S&J programmes, ensure that spending is driven by inclusive national priorities, and are cost-effective and sustainable.
Restructuring the security and justice sector
High order interventions focussed on restructuring the security and justice sector or specific institutions and agencies. This includes structural reforms of existing institutions and interventions to create or establish new organisations and agencies, for example, as part of state-building efforts in fragile or conflict affected contexts where they do not exist or have broken down.
Gendered specific interventions and approaches to reform
This includes efforts towards gender balancing, or ensuring equal representation of men and women in institutions and oversight bodies. This also includes gender-specific interventions such as, VAWG/GBV programming, training and capacity building, creating gender units within the police, and raising awareness of women’s rights within security and justice institutions.
Investment in infrastructure and equipment
Interventions that focus on the provision of physical infrastructure (e.g. buildings, police stations, training centres) or equipment (e.g. law enforcement equipment, uniforms, information technology)
Integrated political engagement/Activities that promote political will
Activities that promote political will, focusing on support that makes the case for change, rather than on-going pressure. This includes activities such as bringing politicians/heads of agencies to meet counterparts in country (e.g. guest of government visits), or sustained lobbying from donor/international community in country, or support to government commissions/task forces.
Demobilisation, disarmament and reintegration (DDR)
DDR typically involves dismantling the command structures of armed groups and reducing the size of fighting forces and the number of weapons in circulation. Ex-combatants are either assisted to return to civilian life, with reintegration packages including cash or non-monetary benefits such as vocational training or counselling, or merged into new national security forces
June 2019
48
Legal services Interventions to provide or broaden the range of legal services available to citizens
Interventions to provide or broaden the range of legal services available to citizens as part of access to justice or legal empowerment programmes. This includes alternative dispute resolution, legal aid, para-legal services, mediation and legal education.
Non-state justice forum and justice facilitator capacity building
Interventions to build the capacity of non-state justice forums (including the full spectrum of traditional and customary justice providers), and local justice facilitators (those individuals working with communities to provide legal aid, dispute resolution, para-legal or mediation services).
Preventative interventions
Integrated efforts to prevent violence and crime, particularly in urban areas. These are often described as integrated development and governance strategies that combine elements of public health, education and employment, voluntary arms collection (demobilisation), community reintegration and urban environmental design programmes.
International mutual legal assistance
Specific mutual legal assistance interventions designed at increasing judicial and legal cooperation across borders. This includes efforts at gathering and exchanging information and evidence, the transferral of proceedings and recognition of judgements.
Displacement and refugee related justice in post-conflict or fragile contexts
Specific justice interventions designed to improve the access of internally or externally displaced persons to justice systems including, but not limited to, in the process of restoration of housing, land and property and in the resolution of human rights abuses relating to displacement.
Outputs
The ultimate aim of this review is explore links between interventions and outcomes, but because it
is anticipated that few studies will rigorously explore these links, any intermediate output and
outcome measures should also be coded.
Outputs Definition
Accountability, effectiveness and transparency
Are the security and justice sectors perceived as effective, accountable and transparent (Global goal 16.6)? Key indicators include: the level of politicisation; the extent of judicial independence; the presence and functioning of internal and external oversight mechanisms; the extent of democratic control; whether security and justice actors behave in a manner accountable to citizens and up to expected standards of behaviour. In addition, indicators could include: primary government expenditures as a proportion of original approved budget for the S&J sector, proportion of the population satisfied with their last experience of S&J service provision.
The inclusion of non-state actors in reform processes or negotiations
Have non-state actors been included as part of reform efforts? This includes the involvement and engagement of legal and statutory actors, (including traditional leaders, civil society organisations and non-governmental organisations), and extra-legal and armed non-state actors in national decision-making processes, agreements, and in the planning and implementation of reform efforts.
Community participation and voice
Have citizens, communities and traditional leaders (including customary justice ‘elders’) had opportunities to participate and voice their opinions and needs with regard to the delivery of security and justice? Key indicators include: whether citizens, communities and traditional leaders
June 2019
49
have been involved in community-based initiatives; whether initiatives have supported the engagement of marginalised and vulnerable groups; whether interactions have occurred between communities, traditional leaders, local officials/ authority/security and justice agency representatives.
Demonstrable capacity of state and non-state organisations to deliver
Have state or non-state actors’ capacities to deliver security and justice been impacted? What is the overall quality of institutions, organisations and customary justice or alternative dispute resolution mechanisms? Key indicators include levels of operational capability, performance, effectiveness, skills, professionalisation and leadership; the ability to plan; the presence of management and performance systems, including recruitment and management and human resource processes. This includes evidence provided on the ability of justice actors to resolve, settle cases and disputes, and the ability to enforce judgements and prosecute cases. It also includes evidence on rates of pre-trial detention.
Citizen confidence, trust or satisfaction in providers
Do citizens or communities report confidence or trust, or are they satisfied with levels of service? This is normally demonstrated through perception surveys or anecdotal evidence. In addition to confidence, trust and satisfaction, a key indicator is whether citizen or communities perceive providers as legitimate. Another aspect of this is whether interventions have led to behaviour change amongst citizens, which means they are more confident of accessing justice
Gender sensitivity and balancing
Have interventions affected the responsiveness of actors and institutions to gender-based violence? Have reform efforts impacted the equal representation of men and women? Key indicators include: perception of how security and justice actors respond to gender-based violence; the implementation of gender related legislation; equal representation through recruitment policies.
Roles, coordination and dialogue amongst organisations and agencies
Are the roles and responsibilities of different actors within the security sector clearly defined? Is there dialogue and coordination amongst different security and justice organisations? Key indicators include: the presence of legislation or frameworks that clearly define the roles and responsibilities of different security and justice actors; the extent of cooperation and lack of conflict between different security and justice actors.
Strategic frameworks developed
Have strategic frameworks been developed by national governments to set out resource allocation and inform implementation decisions? The key indicator is whether frameworks - in the form of strategies, plans, reviews, or legislation – actually exist.
Application, compliance and interpretation of laws
Have legal reforms or the adoption of new codes, norms and standards affected the fair and consistent application and interpretation of law by judges and traditional leaders? This is demonstrated through user perceptions or external analysis of the application of law by different official and non-state justice forums, and whether this is considered to be consistent and fair across different types of cases and for different types of users, especially women and other disadvantaged groups.
State/non-state linkages
Have interventions had any impact upon the level of coherence and linkages between official and customary justice? Have interventions resulted in linkages and coordination between state agencies and non-governmental organisations advocating for security and justice reform? This could be through recognition, regulation, institutionalisation or
June 2019
50
integration of non-state legal orders, including codification. It could also be through the clarification of boundaries between state and non-state jurisdiction, through the diversion of cases from state to non-state forums, or the creation through legislation of hybrid forums
Outcomes
Outcomes Definition
Security and justice actors have incentives for improved service delivery
Has political engagement resulted in political incentives for improved service delivery? The key indicator was where the literature directly referred to impacted incentives for security and justice actors to participate in processes or reforms. This relates to whether there is political or economic benefit for stakeholders to pursue or enact reforms.
Provision is responsive to citizens’ needs
Is provision reflective of citizen concerns and responsive to their needs and priorities? Key indicators include: whether service delivery has been informed by citizen perceptions; the extent to which provision is centred on the needs of citizens and society rather than those of the state or political elites; whether service delivery is responsive to the needs of marginalised and vulnerable groups; whether provision is culturally appropriate; the extent of customer orientation in service delivery; perceptions or evidence of improved service delivery. Have reform efforts improved the responsiveness of actors and institutions to gender-based violence? Have reform efforts increased the equal representation of men and women? Key indicators include: perception of how security and justice actors respond to gender-based violence; the implementation of gender-related legislation; equal representation through recruitment policies.
Political will to enact reforms
Has political engagement resulted in the political will amongst local and national stakeholders to enact reforms or support interventions? Key indicators include: the perceived extent of political will amongst stakeholders; the perceived presence of motivation, commitment or consensus amongst stakeholders for change; the absence of political resistance or spoilers for change.
Ownership of reforms by national and local stakeholders
Do national and local stakeholders demonstrate on-going ownership in the planning or implementation of reforms or interventions? Key indicators include: the extent of perceived ownership by national and local stakeholders; the extent to which reforms are internally initiated by national and local stakeholders; the involvement of a broad range of stakeholders (not just elites) in reform; the extent to which donors take a controlling role in the planning and implementation of reforms.
Citizens feel safe and secure
Do citizens feel safer and more secure during or after interventions have been implemented? This is demonstrated by subjective citizen perceptions or anecdotal accounts of safety and security. This also includes subjective feelings of legal empowerment. This might relate to the perceived ability of citizens to protect their rights or the perception of fewer legal problems in resolving legal conflicts and disputes.
Equal access to provision
Have interventions impacted access to provision for citizens– at a variety of levels, from state to customary? Key indicators include: proportion of victims of violence who reported their victimization to competent authorities or other officially recognized conflict resolution mechanisms, or unsentenced detainees as a proportion of overall prison population.,
June 2019
51
whether provision exists in significant numbers across territories; the extent of access in rural areas; whether cultural, financial or gender related issues prevent access to provision; whether legal services interventions including legal accompaniment have affected access by citizens to the formal justice system (Global Goal 16.3). Other key indicators include the existence of appropriate customary justice or alternative dispute resolution mechanisms, which engage with dispute resolution and adjudication, reparation measures that aim to redress past wrongs and provide compensation or rehabilitation for victims of abuses.
Stability and outbreaks of conflict or violence
Have interventions impacted stability and/or conflict and violence, including local-level conflicts? Have interventions contributed to peacebuilding, political stability and community reconciliation? This is demonstrated by: the absence of political instability or conflict between competing factions and state agencies; and the ability of security and justice agencies to prevent outbreaks of conflict and violence amongst society; and reductions in local-level legal conflicts between disputing parties. Key indicators include recorded instances of violence or political instability between competing factions or cases of local-level community reconciliation.
Resource allocation / funding stability and sustainability
Have reforms contributed to resource allocation, funding stability, or sustainability? Key indicators include: whether funding has been strategically allocated; the absence of resource short-falls; the extent of dependency on donor funding to continue funding the security and justice sectors or reforms after donor interventions have ended.
Security and justice actors are a source of protection, not insecurity
Do security and justice actors, in all of their variety, protect citizens after reforms have been enacted? This is demonstrated by the absence of criminality, grievances, human rights violations, intimidation, exploitation, abuse, extra-legal methods, extremist narratives and violence by security actors against citizens.
Judicial redress to protect rights
Have interventions impacted the ability for citizens to seek redress? Key indicators include the existence of appropriate customary justice or alternative dispute resolution mechanisms, which engage with dispute resolution and adjudication, reparation measures that aim to redress past wrongs and provide compensation or rehabilitation for victims of abuses.
Measures implemented to improve compliance with human rights standards
Have measures been implemented to ensure the compliance of the security and justice actors with international human rights standards? Key indicators include: the adoption of relevant legislation, ethical codes or codes of conduct; human rights training; vetting, certification and removal procedures to ensure that personnel are human rights abiding.
Actual crime rates Have crime rates altered as a result of reforms, or after reforms have been implemented? This is demonstrated through measuring crime in locations where reform initiatives have been implemented.
Legal awareness and confidence
Has the provision of legal services affected awareness amongst citizens of their legal rights? Key indicators include the implementation of activities designed to increase citizen’s knowledge of the law, procedures and available resources for accessing justice, and about how their legal rights are being violated. This could include awareness raising sessions, advocacy and educational campaigns and open forums. The literature may also contain evidence of behaviour change whereby citizens are more confident
June 2019
52
in applying the knowledge and skills they have acquired to access justice or other public services.
Rates of gender-based violence rates
Have interventions had any impact on rates of GBV, including domestic violence? This is demonstrated through explicit analysis on the levels of GBV in locations where initiatives have been implemented.
Local or national economic development
Have reforms contributed to local or national economic development? Key indicators include: impacted GDP; confidence of businesses and entrepreneurs to invest due to security or laws that promote economic activity; the confidence of businesses to operate and/or return to previously insecure locations.
Poverty reduction Have interventions led to poverty reduction? The key indicator is where the evidence presents poverty reduction as an explicit outcome of interventions. This could be through increased incomes or though other human development indicators, such as increased education rates or improved health indicators, where the intervention has taken place.
Access to land, inheritance and property rights
Have interventions affected access to land, inheritance or property rights? Many legal empowerment and access to justice initiatives claim this as a key goal, especially in the case of women and other marginalised groups.
Access to public services and economic resources
Have interventions impacted access to education, health or public utilities, or to services that may help increase incomes, such as employment services, livelihood materials and credit facilities? This could be through making citizens more aware of their rights with regard to accessing public services and/or making public service delivery more responsive.
Women’s empowerment and gender equality
Have interventions influenced gender equality within communities or perceptions of women’s empowerment? This could be through changing attitudes within local communities on the rights of women or their overall standing in relation to men.
Reduction in illicit financial flows
Indicators can include total value of inward and outward illicit financial flows, or proportion of seized small arms and light weapons that are recorded and traced, in accordance with international standards and legal instruments (Global Goal 16.4)
June 2019
53
Appendix 2: Screening Checklist
SCREENING QUESTIONS: TITLE AND ABSTRACT Yes No Unclear
- Make quick judgements to exclude documents based on the following screening questions
- If you cannot exclude, or if you’re unclear after checking against all criteria, then it should be
included for full text screening.
1. Was the map or review published before 2008 (or 2015 for existing
searches)?
If yes then exclude
2. Is the map written in English?
If no then exclude
3. Is security and justice, broadly defined, the thematic focus of the
map?
Security and justice is defined as the “values and goals (e.g. freedom,
fairness, personal safety) as well as to the various institutions established to
deliver them (e.g. defence forces, police, courts). An environment where the
rule of law is respected and security bodies are under the control of civilian
authorities will help people feel safe and secure and encourage them to
claim their rights as citizens. Conversely, where there is no effective and
accountable national security structure, violence can permeate society and
injustice can prevail”6
If no then exclude
4. Is the map a peer-reviewed journal article, edited book chapter,
evaluation, workshop or conference report, working paper,
systematic review or meta-review?
a. Only materials whose primary purpose is to present
empirical evidence will be included. Policy statements,
guidance notes, and advocacy-oriented materials will not be
included. Theoretical or conceptual studies will also not be
included.
If no then exclude
6 DFID., (2007) Explanatory note of security and access to justice for the poor, DFID Briefing April 2007, London: DFID
June 2019
54
5. Is the geographic focus a low- or middle- income country?
If no then exclude
6. Is the research design a primary, empirical research or evaluation, or
a secondary evidence review?
If no then exclude
7. Does this map focus on an INTERVENTION?
a. For our purposes, an intervention could be a local project, a
broader programme or the implementation of a policy. The
intervention may be implemented by any of a range of
actors, including donors, host governments, local
governments, non-governmental organizations and/or local
civil society organizations.
b. Exclude all theoretical or conceptual studies, or those that
mention the impact of any phenomenon other than an
intervention, policy and/or programme
SCREENING QUESTION: FULL TEXT
Repeat questions 4 to 7 and then move onto question 8
8. Does the map have a clear and transparent methodology?
If no then exclude
9. Does the map concern one our included intervention areas? Refer to
Appendix 1 interventions and cross-reference.
If no then exclude
Top Related